Parametric Optimisation of Carbon – Dioxide Moulding Process for Maximum Mold Hardness

Parametric Optimisation of Carbon – Dioxide Moulding Process for Maximum Mold Hardness

M. VenkataRamana Department of MechanicalEngineering

CVR College of Engineering Hyderabad, India

AbstractMould hardness is the important property required for a mould while casting high density alloys. In the present study, it is attempted to optimize the process parameters of CO2 moulding process for maximizing its mould hardness. Four process parameters i.e % of sodium silicate, quantity of CO2 gas, mixing time and % of Coal dust are considered. Taguchi method is employed for the purpose .ANOVA & F-test are conducted to discriminate significant factors from the remaining factors. It is observed that all the considered process parameters are significant for maximizing mould hardness

KeywordsMould Hardness,Optimization,ANOVA

I. INTRODUCTION

Sand molding is a conventional molding process to provide castings of various alloys in to different shapes as per the requirement .Variety of sand moulding process are available

.But still carbon dioxide moulding process has its own merit in producing castings of sound quality.First sodium silicate process was conceived in 1898 and it has become a popular process for core making by 1956.

1.1Carbon Dioxide Moulding Process:-

In CO2 moulding process, moulding sand is mixed with sodium silicate .The mix is loosely rammed in the mould around the pattern. Carbon dioxide gas is sent into the mould

.carbon dioxide is mixed with sodium silicate and hence silica gel is formed.

Na2SiO3 + H2 O + CO2 —– Na2 CO3 + SiO2 + H2O —– (1)

Attrton [1] expressed that there is considerable evidence to show that equation (1) is over simplified and in many ways always incorrect. Nevertheless, silica gel is a new product that is found most often in reactions caused by various hardeners.

Sol i.e colloid of solid in a liquid formed when fine soild particles are dispersed in a liquid. In some cases it is possible to make the dispersed phase coagulate and enclose all the liquid within it. This new product is called gel and in the present case it is silica gel .In silica gel the dispersed phase is silica.

During CO2 moulding process gel formation can be visualized as follows:

Water solution of sodium silicate reacting with CO2 produces mono silisic acid and sodium carbonate.

Na2SiO3+ SiO2+ 2 H2 O + CO2—- Si (OH)4+ Na2 CO3————-(2)

While giving a comprehensive description of strength development process in silicate bonded mould Sarkar[2] mentioned that silicic acid gel with condensed water precedes formation of silica gel and strength of silicic acid gel is achieved by removing the condensed water. Warry [3] described silicic acid gel as loose network of colloidal particles interconnected at only few points in which water and dissolved salts are immobilized .It is prevented from collapse by repulsive forces between other parts of particles. Sarkar [2] mentioned that over -gassing is a consequence of expelling water held in capillarity below a certain level causing gel to shrink and form cracks

Into to it can be concluded that the prime responsible for strength development in silicate bonded moulds is silica gel.

1.2. Importance of present investigation:-

Carbon dioxide moulds are useful for pouring alloys of high density like steel [4] .Hence the requirement of mould in high hardness and strength .High hardness of mould resist penetration of hot liquid metal, usually of high density, in to mould wall. So hardness of mould should be higher the better.

By varying different parameters of moulding process the properties of mould wall will alter. So it is necessary to find out a set of values of parameters of the process that maximizes mould hardness. Previously M.venkataRamana

1. Taguchi method: Conventional full factorial experimentation causes lot of time and money to arrive at optimum process parameter values [6]. It is very difficult to analyze the numerous data obtained and draw proper conclusions.

   Design of experiments is highly useful in planning the experiments systematically and also helps in drawing a meaningful conclusion [7] Several stratified techniques are available to plan for engineering and scientific studies. One such technique is Taguchi method which prescribed a standard way to utilize design of experiment technique for improvement of quality of products and process [8] Several researcher [9,10] across the globe attempted optimization of process parameters of various manufacturing processes.

   1. OBJECTIVE OF PRESENT INVESTIGATION:

      In the present investigation it is attempted to optimize the process parameters of CO2 molding process for maximizing the mould hardness of mould. Taguchi method is employed for the purpose of optimization

   2. SEQUENTIAL STEPS OF INVESTIGATION:

Sequential steps of the present investigation are

1. Identification of process parameters that affect mould hardness and fixing their level values.

2. Deciding suitable experimental plan using Taguchi orthogonal arrays

3. Arrangement of setup for supplying desired quantity of CO2 gas

4. Experimental Determination of mould hardness as per the plan.

5. Analysis of results

3.i .Identification of process parameters that affect mould hardness and fixing their level values:-

Process parameters that affect mould hardness of CO2 moulding process are:

1. Percentage of sodium silicate

2. Quantity of CO2 gas

3. Mixing time of sodium silicate and sand mix

4. Percentage of coal dust.

3.i.a)percentage of sodium silicate : Sodium silicate is prime responsible for strength and hardness of mould made of CO2moulding process .Usually 3% to6% sodium silicate is used in the process [11].Hence in present investigation % of sodium silicate is considered at two levels i.e 4% and 6%.

3. i.b) Quantity of CO2gas( CO2 gassing time): Carbon dioxide gas , when passed through the mould, mixes with sodium silicate and silica gel is formed. This silica gel is the main bonding agent between the sand grains .If sufficient quantity of CO2 gas is not supplied it may hamper the bonding process. At the same time if more amount of CO2gas is supplied, it may result in to formation of white patches indicating the formation of sodium bicarbonate .This sodium bicarbonate formation reduces the strength of the bond. Hence adequate supply of CO2gas is essential for maximizing mould hardness. As per the Chemical balance between CO2 and Na2 O of Sodium silicate, it is estimated that 10kg of CO2gas is required for 100kg of silicate used. However in shop floor conditions, quantity up to 40kg of CO2 per 100kg of sodium silicate is used

The two levels of CO2gas supply are appropriately converted into gassing time and hence in this paper the parameter, quantity of CO2 gas is termed as gassing time. First level of quantity of CO2gas i.e 10kg per100kg of sodium silicate yield to 13 seconds gassing time and second level i: e 40kg per100kg of sodium silicate yields to 30 seconds.

3. i.c)Mixing time: Mixing of sand and sodium silicate ensures uniform coating of silicate on sand grains which will further ensure proper formation of silica gel on the entire sand grain surface. Generally a mixing time of 5 minutes to 10minutes is employed .In present study two levels i.e. 5 minutes and 10minutes are considered.

3.i.) Coal dust: Coal dust addition enhances collapsibility of CO2 mould after pouring .Usually 2% of coal dust addition are made [12] .The two levels of coal dust considered are 0% and2%.

Process parameters and their level values are given in Table-1 Table-1 Description of factors and their level)

3. ii)Deciding suitable experiment plan using Taguchi orthogonal array.

Four factors at two levels lead to 4 degrees of freedom and the two interactions between sodium silicate & gassing time and sodium silicate & mixing time lead to two more degrees of freedom and total degrees of freedom account to 6 .So a L8 orthogonal array with seven degrees of freedom suffices the requirement.

L8 orthogonal array with actual level values of factors is presented in Table2

Table -2 Experimental plan

1. Iii) Arrangement of setup for supplying desired quantity of CO2 gas:

   To ensure adequate supply of CO2 gas a gassing arrangement setup with Pressure gauge and Rota meter is made.

   1. iv)Experimental Determination of mould hardness as per the plan.

      Fresh silica sand with a grain Fineness number of 59.87 is considered in present work. Screen distribution of the used sand is given Table-3

      Table-3 Screen distribution of fresh silica sand

      Aperture Opening	Wt.of the sand retained in each sieve(gm)	Percentage retained	Multiplier	Product
      1700	–	–	5	–
      850	0.8	1.6	10	16
      600	1.9	3.8	20	76
      425	9.3	18.6	30	558
      300	10.8	21.6	40	864
      212	11.9	23.8	50	1190
      150	5.8	11.6	70	812
      106	4.6	9.2	100	920
      75	2.1	4.2	140	588
      53	1.6	3.2	200	640
      Pan	0.4	0.8	300	240
      Total	49.3	98.6		5904

      Table-4 : Experimental values of mould hardness

      Expt Trail No	Mould hardness (Scratch Hardness No)			Average Value	S/N ratio
      1	72	72	71	71.67	37.105
      2	63	64	67	64.67	36.204
      3	75	71	70	72	37.135
      4	71	75	72	72.67	37.219
      5	81	79	81	80.33	38.096
      6	75	78	80	77.67	37.795
      7	81	82	82	81.67	33.24
      8	78	78	76	77.33	37.765

      Quality characteristic for mould hardness is Bigger the better type

      S/N ratio = -10 log(MSD) ——-(3) Where MSD=mean square deviation

      For Bigger the better typequality characteristic

      MSD = [(1/y12+1/y 2+1/y 2+.]/n—————-(4)

      AFS standard specimen of 2 X2 cylindrical shape are prepared as per experiment plan mentioned in Table-2.Mould hardness values are determined using scratch hardness tester. Standard AFS specimen and scratch hardness tester are shown in Fig-1 and Fig-2 respectively.

      Fig.1 AFS standard sand specimen

      Fig.2 Scratch hardness tester

      Experimentally determined mould hardness values of fresh silica sand mould are tabulated in Tab-4

      2 3

      3.v Analysis of Results

      37.59

      37.59

      37.64

      37.4537.44

      37.64

      37.4537.44

      37.59

      37.59

      37.4

      37.4

      37.32

      37.32

      37.3

      37.3

      37.3

      37.3

      37.2

      37.2

      37.24

      37.24

      S/N Ratio

      S/N Ratio

      S/N analysis of experimentally obtained mould hardness values are made and response graphs are shown Fig-3

      38

      37.97

      38

      37.97

      37.8

      37.8

      37.6

      37.56

      37.6

      37.56

      37

      37

      36.91

      36.91

      36.8

      36.8

      0 1 2 3 4 5 6 7 8 9 10 11 12

      0 1 2 3 4 5 6 7 8 9 10 11 12

      SS

      GT

      SSXGT

      MT

      SSXMT

      CD

      SS

      GT

      SSXGT

      MT

      SSXMT

      CD

      Fig.3:Response graph

      Optimum condition for maximizing mould hardness can be determined based on response graphs, but prior to that significant and insignificant factors are to be discriminated. This can be done by performing Analysis of Variance (ANOVA) on obtained mould hadness data and performing

      F-Test

      . Analysis of Variance for Mould hardness is given in Table- 5.

      Tables-5 : Analysis of Variance

      After obtaining optimum value of S/N ratio MSD can be calculated using equation 3 and by using equation 4, MSD value can be converted in to actual value of mould hardness and the value is 83.76

      3.v.c)Range of Expected mould hardness at optimum condition

      Range of expected mould hardness at the obtained optimum condition of process parameters with specified confidence level is determined as follows

      Confidence level(CL)=95% F(n1,n2)=F(1,1)=1.6

      Confidence interval (CI)

      Factor Name	DOF	SS	Variance	F-ratio	Percentage of contribution of factor
      SS	1	2.239	2.239	27.816	72.6
      GT	1	0.167	0.167	2.08	5.418
      SSXGT	1	0.108	0.108	1.353	3.504
      MT	1	0.317	0.317	3.946	10.28
      SSXMT	1	0	0	0	0
      CD	1	0.167	0.167	1.98	5.418
      Error	1	0.084	0.084		2.78
      Total	7	3.082			100

      Factor Name	DOF	SS	Variance	F-ratio	Percentage of contribution of factor
      SS	1	2.239	2.239	27.816	72.6
      GT	1	0.167	0.167	2.08	5.418
      SSXGT	1	0.108	0.108	1.353	3.504
      MT	1	0.317	0.317	3.946	10.28
      SSXMT	1	0	0	0	0
      CD	1	0.167	0.167	1.98	5.418
      Error	1	0.084	0.084		2.78
      Total	7	3.082			100

      CI =

      F(1, n2)*Ve / NE

      ———– (5)

      3.v.a)F-Test : From F-test it can be observed that except interactions all the factors are significant. Hence from response graph and F-Test optimum condition can be obtained and the obtained condition is given inTable-6 Percentage contribution of each factor and interaction towards mould hardness is given in Fig-4

      % contrbution of Factors

      % contrbution of Factors

      Table-6: Optimum condition

      Factor Name	Level Description	Level Contribution
      SS	6%(2)	0.529
      GT	30 sec(2)	0.144
      SSXGT	(2)	0.116
      MT	5 min (1)	0.199
      SSXMT	(1)	0.005
      CD	0% (1)	0.144

      100

      80 72.6

      60

      40

      20

      5.418

      3.504

      10.28

      0

      5.418

      2.78

      100

      80 72.6

      60

      40

      20

      5.418

      3.504

      10.28

      0

      5.418

      2.78

      0

      0

      SS

      GT SSXGT MT SSXMT CD Error

      SS

      GT SSXGT MT SSXMT CD Error

      Fig-4Percentage contribution of each factor and interaction on mould hardness

      Grand average of S/N values of mould hardness (T )=37.445 3.v.b)Expected mould hardness at optimum condition: Expected mould hardness at optimum condition

      Y optimum = T + [SS 1 – T ] + [ G T2 – T ] + [ MT1 – T ] +[CD1

      – T ]

      = 37.445+0.529+0.144+0.199+0.144

      =38.461 (S/N value) (based on the significance of factors and their percentage contribution towards result)

      Ve =Error variance=0.084 Ne=Effective number of replications

      =Total number of S/N values

      ————————————-

      DOF of mean (always=1)+DOF of factors

      =8/(1+4)=1.6

      Confidence interval (C.I)=+/- 0.2828

      Range of Mould hardness at optimum condition = (38.461- 0.2828) to (38.461+0.2828)=38.1782 to 38.743(S/N value)

      Using equation 3 and4 S/N values are converted in to actual values of mould hardness and it is found that the range of mould hardness value at optimum condition is from 81 to 86.5

      Conformation test is conducted at obtained optimum condition and average of three mould hardness values at this optimum condition is observed to be 82. This mould hardness value is well within the range of expected mould hardness value at optimum condition

   2. CONCLUSIONS:

Using Taguchi Technique Process parameters of CO2 moulding process are successfully optimized for maximum mould hardness. All the process parameters considered i.e % of sodium silicate, quantity of CO2 gas mixing time and % of coal dust are observed to be significantly affecting mould hardness Optimum value of process parameters are

% of Sodium Silicate: 6% ; Quantity of CO2 gas (gassing time):30secnds

Mixing time: 5 minutes; % of Coal dust:0%

None of the considered interactions are found to be significant. Mould hardness obtained through validation experiment conducted at optimum condition is well within the expected range of mould hardness at optimum condition.

REFERENCES

1. D.V Atterton, The Carbon-Di-Oxide Process,AFS Transaction Vol.64,pp-14-18 (1956)

2. A.D Sarkar ,Foundry Core and Mould Making ,U.K Pergaman press (1964)

3. CIATF Development of a Method of Reclamation of Self-Setting Fluidized Sands,AFS International Cast Metals Journal pp 37-40,Dec 1978

4. Principles of Metal Casting-RWH Hooper & Rosenthall ,TataMC graw hill education

5. M.Venkata Ramana Optimization of process parameters of CO2 moulding process for better Tensile strength-Taguchi approach,International journal of Applied Engineering Research,vol.10,No.11(2015) pp28145-28154

6. Ros Philip J Taguchi Technique for Quality Engineering,M.C.Grawhill Book company,Newyork(1996)

7. G.Taguchi,Introduction to Quality Engineering,Asian Productivity organization ,Tokyo(1990)

8. Ranjith K.Roy. Design of Experiments using Taguchi ApproachProceedings of work shop on Taguchi technique, JNTU,Hyderabad(2003)

9. J.L .Lin K.S Wang,B.H Yan and Y.S .Tarang Optimization of Electrical discharge machining process based on Taguchi method with

   Fuggy logic,Journal of Meterial Processing Technology ,Vol.102,pp 48-55 (2000)

10. P. B Barua ,P.Kumar and J.L Gaindha,Surface Roughness Optimization of V-Process Casting through Taguchi Methos,AFS Transactions ,Vol 105,pp 763-768 (1997)

11. K.Srinagesh, Chemistry of Sodium Silicate as a Sand Binder,Cast Metals Research Journal,pp-50-63 (March 1979)

12. A.R.Krishnamoorthy Sand Practices ,The institute of Indian Foundrymen,Calcutta(1996).

.